Because we have previously observed substantial responses to deoxycoformycin in hairy cell leukemia after failure of interferon, we began this pilot protocol to evaluate the efficacy and toxicity of combined recombinant interferon alfa-2a and 2'deoxycoformycin. Response criteria in this protocol were more strict than other studies in that we required absence of hairy cells from bilateral iliac crest bone marrow biopsies obtained twice at approximately two month intervals and improvement of the peripheral blood hemoglobin to greater than 12 grams percent, granulocyte count to greater than 1500 per cubic millimeter, and platelet count to greater than 100,000 per cubic millimeter. Most previous studies using deoxycoformycin have declared patients complete responders on the basis of a single unilateral bone marrow biopsy. Of 15 patients, 13 are evaluable for response with one patient being inevaluable because of a diffuse bony sclerosis and consequent inability to provide marrow for assessment of the marrow response, and one patient continues on therapy. All 13 patients had partial responses with small numbers of residual hairy cells remaining in the bone marrow. Had standard criteria for complete remission been used, we would have had 10 complete responses based on only a single unilateral bone marrow biopsy. It appears that the combination of deoxycoformycin plus interferon provides no substantial improvement in complete response rate over dCF alone. At present, therefore, the combination of deoxycoformycin and recombinant interferon alfa-2a cannot be recommended over either agent alone.